Vitaly Lavrukhin

发表

Training Deep Networks with Stochastic Gradient Normalized by Layerwise Adaptive Second Moments

O. Kuchaiev, Boris Ginsburg, P. Castonguay, 2019 .

Damage Control During Domain Adaptation for Transducer Based Automatic Speech Recognition

Boris Ginsburg, Somshubra Majumdar, Vitaly Lavrukhin, 2022, 2022 IEEE Spoken Language Technology Workshop (SLT).

Jasper: An End-to-End Convolutional Neural Acoustic Model

Boris Ginsburg, Oleksii Kuchaiev, Huyen Nguyen, 2019, INTERSPEECH.

Mixed-Precision Training for NLP and Speech Recognition with OpenSeq2Seq

Boris Ginsburg, Oleksii Kuchaiev, Paulius Micikevicius, 2018, 1805.10387.

OpenSeq2Seq: Extensible Toolkit for Distributed and Mixed Precision Training of Sequence-to-Sequence Models pdf

Boris Ginsburg, Oleksii Kuchaiev, Paulius Micikevicius, 2018, ArXiv.

Quartznet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions

Boris Ginsburg, Oleksii Kuchaiev, Yang Zhang, 2019, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

Citrinet: Closing the Gap between Non-Autoregressive and Autoregressive End-to-End Models for Automatic Speech Recognition

Boris Ginsburg, Somshubra Majumdar, Oleksii Hrinchuk, 2021, 2104.01721.

Cross-Language Transfer Learning and Domain Adaptation for End-to-End Automatic Speech Recognition

Boris Ginsburg, Jing Xiao, Jianzong Wang, 2021, 2021 IEEE International Conference on Multimedia and Expo (ICME).

NeMo: a toolkit for building AI applications using Neural Modules pdf

Boris Ginsburg, Oleksii Kuchaiev, Jason Li, 2019, ArXiv.

SpeakerNet: 1D Depth-wise Separable Convolutional Network for Text-Independent Speaker Recognition and Verification.

Boris Ginsburg, Nithin Rao Koluguri, Jason Li, 2020, 2010.12653.

Cross-Language Transfer Learning, Continuous Learning, and Domain Adaptation for End-to-End Automatic Speech Recognition

Boris Ginsburg, Oleksii Kuchaiev, Patrick O'Neill, 2020, 2005.04290.

Training Neural Speech Recognition Systems with Synthetic Speech Augmentation pdf

Boris Ginsburg, Jason Li, Vitaly Lavrukhin, 2018, ArXiv.

Hi-Fi Multi-Speaker English TTS Dataset

Boris Ginsburg, Yang Zhang, Evelina Bakhturina, 2021, Interspeech.

SPGISpeech: 5, 000 hours of transcribed financial audio for fully formatted end-to-end speech recognition pdf

Shinji Watanabe, Boris Ginsburg, Somshubra Majumdar, 2021, Interspeech.

Improving Noise Robustness of an End-to-End Neural Model for Automatic Speech Recognition

Boris Ginsburg, Somshubra Majumdar, Jagadeesh Balam, 2020, 2010.12715.

Stochastic Gradient Methods with Layer-wise Adaptive Moments for Training of Deep Networks pdf

Boris Ginsburg, Oleksii Kuchaiev, Huyen Nguyen, 2019, ArXiv.

A Toolbox for Construction and Analysis of Speech Datasets

Boris Ginsburg, Evelina Bakhturina, Vitaly Lavrukhin, 2021, NeurIPS Datasets and Benchmarks.

Text-only domain adaptation for end-to-end ASR using integrated text-to-mel-spectrogram generator

Boris Ginsburg, Vitaly Lavrukhin, Vladimir Bataev, 2023, INTERSPEECH 2023.

NeMo Toolbox for Speech Dataset Construction

Boris Ginsburg, Evelina Bakhturina, Vitaly Lavrukhin, 2021, ArXiv.

NeMo Toolbox for Speech Dataset Construction pdf

Boris Ginsburg, Evelina Bakhturina, Vitaly Lavrukhin, 2021, ArXiv.

LibriSpeech-PC: Benchmark for Evaluation of Punctuation and Capitalization Capabilities of End-to-End ASR Models

Boris Ginsburg, Evelina Bakhturina, Vitaly Lavrukhin, 2023, 2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU).